cpsc 881: information retrieval. 2 hardware basics many design decisions in information retrieval...
Post on 22-Dec-2015
215 Views
Preview:
TRANSCRIPT
CpSc 881: Information Retrieval
2
Hardware basics
Many design decisions in information retrieval are based on hardware constraints.
We begin by reviewing hardware basics that we’ll need in this course.
3
Hardware basics
Access to data is much faster in memory than on disk. (roughly a factor of 10)
Disk seeks are “idle” time: No data is transferred from disk while the disk head is being positioned.
To optimize transfer time from disk to memory: one large chunk is faster than many small chunks.
Disk I/O is block-based: Reading and writing of entire blocks (as opposed to smaller chunks). Block sizes: 8KB to 256 KB
Servers used in IR systems typically have several GB of main memory, sometimes tens of GB, and TBs or 100s of GB of disk space.
Fault tolerance is expensive: It’s cheaper to use many regular machines than one fault tolerant machine.
4
Some stats (ca. 2008)
symbol statistic value
sb
P
average seek timetransfer time per byteprocessor’s clock rateLow level operation (e.g., compare & swap a word)size of main memorysize of disk space
5 ms = 5 × 10−3 s0.02 μs = 2 × 10−8 s109 s−1
0.01 μs = 10−8 s
several GB1 TB or more
5
Blocked Sort-Based Indexing (BSBI) algorithm: construct the
inverted Index
dictonary postings
6
Index construction in IIR 1: Sort postings in memory
7
Sort-based index construction
As we build index, we parse docs one at a time.
The final postings for any term are incomplete until the end.
At 10–12 bytes per postings entry, we need a lot of space for large collections.
T = 100,000,000 in the case of RCV1: we can do this in memory on a typical machine in 2010.
Can we keep all postings in memory and then do the sort in-memory at the end?
Not for large collectionsIn-memory index construction does not scale for large collections.Thus: We need to store intermediate results on disk.
8
Same algorithm for disk?
Can we use the same index construction algorithm for larger collections, but by using disk instead of memory?
No: Sorting T = 100,000,000 records on disk is too slow – too many disk seeks.
We need an external sorting algorithm.
9
“External” sorting algorithm(using few disk seeks)
We must sort T = 100,000,000 non-positional postings.
Each posting has size 12 bytes (4+4+4: termID, docID, document frequency).
Define a block to consist of 10,000,000 such postings
We can easily fit that many postings into memory.
We will have 10 such blocks for RCV1.
Basic idea of algorithm:For each block: (i) accumulate postings, (ii) sort in memory, (iii) write to disk
Then merge the blocks into one long sorted order.
10
Merging two blocks
11
Blocked Sort-Based Indexing (BSBI)
(i) segments the collection into parts of equal size,
(ii) sorts the termID–docID pairs of each part in memory,
(iii) stores intermediate sorted results on disk, and
(iv) merges all intermediate results into the final index.
12
Blocked Sort-Based Indexing
Key decision: What is the size of one block?
13
Problem with sort-based algorithm
Our assumption was: we can keep the dictionary in memory.
We need the dictionary (which grows dynamically) in order to implement a term to termID mapping.
Actually, we could work with term,docID postings instead of termID,docID postings . . .
. . . but then intermediate files become very large. (We would end up with a scalable, but very slow index construction method.)
14
Single-pass in-memory indexing
Abbreviation: SPIMI
Key idea 1: Generate separate dictionaries for each block – no need to maintain term-termID mapping across blocks.
Key idea 2: Don’t sort. Accumulate postings in postings lists as they occur.
With these two ideas we can generate a complete inverted index for each block.
These separate indexes can then be merged into one big index.
15
SPIMI-Invert
16
SPIMI: Compression
Compression makes SPIMI even more efficient.Compression of terms
Compression of postings
17
Distributed indexing
For web-scale indexing (don’t try this at home!): must use a distributed computer cluster
Individual machines are fault-prone.Can unpredictably slow down or fail.
How do we exploit such a pool of machines?
18
Google data centers (2007 estimates; Gartner)
Google data centers mainly contain commodity machines.
Data centers are distributed all over the world.
1 million servers, 3 million processors/cores
Google installs 100,000 servers each quarter.
Based on expenditures of 200–250 million dollars per year
This would be 10% of the computing capacity of the world!
If in a non-fault-tolerant system with 1000 nodes, each node has 99.9% uptime, what is the uptime of the system (assuming it does not tolerate failures)?
Answer: 63%
Suppose a server will fail after 3 years. For an installation of 1 million servers, what is the interval between machine failures?
Answer: less than two minutes
19
Distributed indexing
Maintain a master machine directing the indexing job – considered “safe”
Break up indexing into sets of parallel tasks
Master machine assigns each task to an idle machine from a pool.
20
Parallel tasks
We will define two sets of parallel tasks and deploy two types of machines to solve them:
Parsers
Inverters
Break the input document collection into splits (corresponding to blocks in BSBI/SPIMI)
Each split is a subset of documents.
21
Parsers
Master assigns a split to an idle parser machine.
Parser reads a document at a time and emits (term,docID)-pairs.
Parser writes pairs into j term-partitions.
Each for a range of terms’ first lettersE.g., a-f, g-p, q-z (here: j = 3)
22
Inverters
An inverter collects all (term,docID) pairs (= postings) for one term-partition (e.g., for a-f).
Sorts and writes to postings lists
23
Data flow
24
MapReduce
The index construction algorithm we just described is an instance of MapReduce.
MapReduce is a robust and conceptually simple framework for distributed computing . . .
. . .without having to write code for the distribution part.
The Google indexing system (ca. 2002) consisted of a number of phases, each implemented in MapReduce.
Index construction was just one phase.
25
Index construction in MapReduce
26
Dynamic indexing
Up to now, we have assumed that collections are static.
They rarely are: Documents are inserted, deleted and modified.
This means that the dictionary and postings lists have to be dynamically modified.
27
Dynamic indexing: Simplest approach
Maintain big main index on disk
New docs go into small auxiliary index in memory.
Search across both, merge results
Periodically, merge auxiliary index into big index
Deletions:Invalidation bit-vector for deleted docs
Filter docs returned by index using this bit-vector
Auxiliary index: index construction time is O(T2) as each posting is touched in each merge.
28
Issue with auxiliary and main index
Frequent merges
Poor search performance during index merge
Actually:Merging of the auxiliary index into the main index is not that costly if we keep a separate file for each postings list.Merge is the same as a simple append.But then we would need a lot of files – inefficient.
Assumption for the rest of the lecture: The index is one big file.
In reality: Use a scheme somewhere in between (e.g., split very large postings lists into several files, collect small postings lists in one file etc.)
29
Logarithmic merge
Logarithmic merging amortizes the cost of merging indexes over time.
→ Users see smaller effect on response times.
Maintain a series of indexes, each twice as large as the previous one.
Keep smallest (Z0) in memory
Larger ones (I0, I1, . . . ) on disk
If Z0 gets too big (> n), write to disk as I0
. . . or merge with I0 (if I0 already exists) and write merger to I1 etc.
30
Logarithmic merge
31
Logarithmic merge
Number of indexes bounded by O(log T) (T is total number of postings read so far)
So query processing requires the merging of O(log T) indexes
Time complexity of index construction is O(T log T).. . . because each of T postings is merged O(log T) times.
So logarithmic merging is an order of magnitude more efficient.
32
Dynamic indexing at large search engines
Often a combinationFrequent incremental changes
Rotation of large parts of the index that can then be swapped inOccasional complete rebuild (becomes harder with increasing size – not clear if Google can do a complete rebuild)
33
Building positional indexes
Basically the same problem except that the intermediate data structures are large.
34
Take-away
Two index construction algorithms: BSBI (simple) and SPIMI (more realistic)
Distributed index construction: MapReduce
Dynamic index construction: how to keep the index up-to-date as the collection changes
top related